ABSTRACT Avoidant and restrictive eating is common across youth mental health disorders, and is prospectively associated with poor growth, nutritional deficiencies, psychiatric symptoms, and psychosocial impairment. The newly introduced DSM-5 Avoidant/Restrictive Food Intake Disorder includes some (but not all) avoidant and restrictive eating presentations. Initial evidence suggests that avoidant and restrictive eating is highly heterogeneous, with poor intake characterized by choking or vomiting phobias; loss of interest in feeding; and/or extreme food selectivity. Moreover, very little is known about its pathophysiology or long- term outcomes. Our study will leverage unique and complementary contributions of an international team with expertise in clinical psychology, child psychiatry, pediatrics, endocrinology, biostatistics and functional neuroimaging to investigate risk mechanisms and longitudinal outcomes of avoidant and restrictive eating. We will establish a clinical cohort of children aged 8-18 years (n=100) and healthy controls (n=50) matched for sex, age, and Tanner stage to investigate how, across units of analysis, RDoC constructs contribute to avoidant and restrictive eating. First, we hypothesize that Negative Valence (acute threat/fear) over-activity (circuits: amygdala activation during a validated fear paradigm; hormones: cortisol and oxytocin; physiology: heart rate and skin conductance; self report: fear and trait anxiety) will correlate with phobic features. Second, we hypothesize that Arousal/Regulatory (homeostasis) dysfunction (circuits: hypothalamus and insula hypoactivation during a validated food paradigm; hormones: PYY, CKK, BDNF; self report: hunger and fullness) will correlate with low-appetite features. Third, we will explore Cognitive Systems (perception) over-sensitivity (circuits: hyperactivation in primary taste cortex during a taste paradigm; hormones: PYY; self report: sensory profile; behavior: taste and odor threshold, discrimination, and detection; physiology: taste perception) and its correlation with sensory features. We expect that the clinical cohort will have greater dysfunction across these 3 constructs than controls. We will then use latent class factor models to determine whether avoidant/restrictive eating comprises multiple distinct phenotypes (as prior literature assumes) or a single phenotype with 3 overlapping dimensions (as we hypothesize). We will follow our clinical cohort for 2 years to evaluate a) the persistence of avoidant/restrictive eating, growth, and psychopathology outcomes; and b) if dysfunction in all 3 RDoC constructs predicts outcomes. This study will be innovative and unique in 3 ways: 1) by providing an empirical investigation of an understudied clinical presentation; 2) by providing the first investigation of pathophysiology and risk mechanisms; and 3) by characterizing poorly understood longitudinal outcomes. In sum, conceptualizing avoidant/restrictive eating within an RDoC framework that integrates mind, brain, and behavior has great potential to reduce the burden of feeding disorders by informing nosology, treatment, and prevention.